Am looking for some test users of an identification app

I wish to start in the middle, actually. The very old ones tend to be things that are completely unidentifiable. I thought I might go back to the website and mark all the first pages reviewed to help with this problem, but turns out there are too many pages in North America for that to be feasible.

1 Like

You could go to page 2 and then edit the page number to be roughly halfwayā€¦

1 Like

There is an error message preventing you from going past page 333 (or page 50 with 200 obs per page) because of something about showing past 10,000 observations.

2 Likes

This is in a mobile app made by @jtklein where there isnā€™t any pagination like that.

Yes I meant in the app. But for the website @kiwifergusā€™s suggestion is useful.

I like your original suggestion of selecting a date range. Or maybe the app should simply default to pulling observations more than a month old.

2 Likes

One thing I realized about this app is the enormous potential for trolling. Those of us who have it now are hopefully serious and trustworthy users of iNaturalist, but if it were widely available for download, thereā€™s nothing stopping a bored malicious person from swiping a few thousand observations into incorrect categories on purpose. I donā€™t know how popular @jtklein imagined the app being, but perhaps iNaturalist the organization would prefer it not spread too much.

2 Likes

ā€œWith great power comes great responsibility.ā€ ;)

5 Likes

I just found one unexpected effect of using this tool: my id count went up markedly enough recently that I think another user presumed I was increasing the count, just to increase my count? Under those circumstances, Iā€™ll probably rethink my idea to go back over a lot of my Plantae labels later on, and just leave them at Plantae. All of those potential changes would wildly increase my count even more. Oh well, itā€™s likely a very rare negative.

I found out today you can also search for things with no ID using https://www.inaturalist.org/observations/identify?identified=false NOTE that link as is loads veeeeeeeerrrry slowly because it returns too many results, and before it loads it will tell you there are no results (itā€™s lying.) If you want it to load faster, you can add into that URL a place with a smaller number of unknowns.

3 Likes

Sorry, somehow I did not get notifications about new posts in this thread here.

About the trolling aspect:
For me it was about speeding up the id process. Yes, you are right if someone wants to make wrong ids it is also faster for that.
However, if someone really wants to troll everyone there are much easier way than swiping on hundreds of observations in my app. You can do that on thousands of observations with one time hitting return if you know how to.

Update: Would you believe there are <800 Unknowns left from Feb ~23 2019 Penang? It was doable to get that far thanks to this tool! :) (Of course most of them are just labeled ā€œPlantaeā€ now but thatā€™s better than it was.)

Meanwhile, one possible feature request based on this post
https://forum.inaturalist.org/t/amount-of-unknown-records-is-decreasing/8594/336

If the app had a toggle setting for Captive (ideally with default = No), that would cause the records mentioned in that post to be handled easily through this tool.

3 Likes

Iā€™d also like to request a new setting, sort by random. If itā€™s not too complicated :)
Iā€™ve been using the app a great deal lately. Itā€™s responsible for maybe 20-25k of my IDs since Juneā€¦ x_x

1 Like

Making multiple IDs on a single observation counts as a single ID. It used to be that refining your ID three times counted as three IDs in the bad old days of the prominently displayed leaderboards, but it hasnā€™t worked like that in years now.

1 Like

Great news, thanks! I can at least do monocot/dicot etc for a lot of those later in that case.

While I am making feature requests, here is one more request for a chunk of Unknowns not yet accessed by the tool:

Include an additional Location choice for ā€œNone of theseā€ or ā€œOtherā€: that would cover the records that are either Private setting (a large number- this would help a lot); or in part of the oceans/an island not covered by the listed geographic regions (a tiny number); or have their location set inaccurately-overbroadly (a tiny number that is being cleaned up otherwise).

1 Like

Ok great, thanks for the request. You can toggle an observation as captive/cultivated in the menu that opens through the round button at the bottom.
Or did you mean a toggle to filter the list of observations for ones which are marked as captive?

I have to investigate how to do the ā€œOtherā€ location setting.

1 Like

That one. :) Thanks again for such a useful tool!

I have tried the new true/false Captive setting (thanks!). I have found that for each continent, I could identify the first set of records. But when trying to retrieve the next set, it spins a little while and then reports an error message.

Thank you for the report. I will try to look into it, to see why it fails.
I have also added a field where you can directly search for the place to filter. Like on the iNat homepage you can select all places now.

5 Likes